# Determine number of entries per gene
gene = colicogs[colicogs['gene_name'].str.lower()==g.lower()]
b_number = gene['b_number'].unique()[0]
gene_product = gene['gene_product'].unique()[0]
go_term = ';'.join(list(gene['go_terms'].unique()))
if len(gene) > 0:
cog_class = gene['cog_class'].values[0]
cog_cat = gene['cog_category'].values[0]
cog_letter = gene['cog_letter'].values[0]
gene_product= gene['gene_product'].values[0]
mw = gene['mw_fg'].values[0]
go_term = ';'.join(list(gene['go_terms'].unique()))
for _c, _d in d.groupby(['growth_rate_hr-1']):
# volume predictions based on MG1655 data, Si, F. et al. (2017, 2019)
vol = size.lambda2size(_c)
# extract relevant information.
gene_dict = {
'gene_name': g.lower(),
'b_number': b_number,
'condition': _d['condition'].unique()[0],
'corrected_volume': vol,
'reported_tot_per_cell': _d['copy_number_molecule-per-fL'].values[0] * vol,
'reported_fg_per_cell': _d['copy_number_molecule-per-fL'].values[0] * vol * mw,
'go_terms':go_term,
'cog_class': cog_class,
'cog_category': cog_cat,
'cog_letter': cog_letter,
'gene_product': gene_product,
'growth_rate_hr': _c
}
# plot the max for respiration
ax1.plot(Ps_resp_rod, SA_V_ratio_rod, color=colors['blue'],
label='rod', alpha=0.9, lw = 0.5, ls = '-.')
ax1.plot(Ps_resp_sphere, SA_V_ratio_sphere, color=colors['blue'],
label='sphere', alpha=0.9, lw = 0.5, ls = '--')
ax1.fill_between(Ps_resp_, y1 = SA_V_ratio_sphere_, y2 = SA_V_ratio_rod_,
color=colors['blue'],alpha=0.2, lw = 0)
# # Populate second plot with growth rates
# S/V for E. coli datasets
# Load the data set
data = pd.read_csv('../../data/compiled_absolute_measurements.csv')
for g, d in data.groupby(['dataset', 'condition', 'growth_rate_hr']):
V = size.lambda2size(g[2])
# ATP equivalents demand w.r.t. volume ; 1E6 ATP/(um3 s)
Pv = 1E6 * V
# assume aspect ratio of 4 (length/width), which is
# appoximately correct for E. coli
SA_rod = 2 * np.pi * V**(2/3)
SV = SA_rod/V
ax1.plot(Pv, SV, 'o', color=dataset_colors[g[0]],
alpha=0.75, markeredgecolor='k', markeredgewidth=0.25,
label = g[2], ms=4, zorder=10)
# Format the axes
for a in [ax1]:#,ax2]:
a.xaxis.set_tick_params(labelsize=5)
a.yaxis.set_tick_params(labelsize=5)
gr_schmidt = d_schmidt_.growth_rate_hr.values[0]
cond_schmidt = d_schmidt_.condition.values[0]
d_schmidt_ = d_schmidt_[d_schmidt_.growth_rate_hr == gr_schmidt]
d_schmidt_ = d_schmidt_[d_schmidt_.condition == cond_schmidt]
schmidt_genes = d_schmidt_[d_schmidt_.b_number.isin(d.b_number.unique())]
rel_schmidt = schmidt_genes.fg_per_cell.sum() / d_schmidt_.fg_per_cell.sum(
)
rel_corr_fg = df[df['growth_rate_hr'] == g]['reported_fg_per_cell'].sum() / \
size.lambda2P(g)
print(g, ': total mass fg: ', np.round(size.lambda2P(g), 2), ' volume: ',
np.round(size.lambda2size(g), 2), ' relative change in total fg: ',
np.round(1 / rel_corr_fg, 2), ' abundance relative to Schmidt: ',
np.round(rel_schmidt, 2))
df.loc[df['growth_rate_hr']==g, 'tot_per_cell'] = \
(df.loc[df['growth_rate_hr']==g]['reported_tot_per_cell'] / rel_corr_fg) * rel_schmidt
df.loc[df['growth_rate_hr']==g, 'fg_per_cell'] = \
(df.loc[df['growth_rate_hr']==g]['reported_fg_per_cell'] / rel_corr_fg) * rel_schmidt
#%%
df['dataset'] = 'valgepea_2013'
df['dataset_name'] = 'Valgepea et al. 2013'
df['strain'] = 'MG1655'
df.to_csv('../../../data/valgepea2013_longform_annotated.csv')
# %%
# Load the complex subunit counts.
subunits = pd.read_csv('../../data/compiled_annotated_complexes.csv')
# # Load the compiled data
data = pd.read_csv('../../data/compiled_absolute_measurements.csv')
# Compute the minimum number of complexes.
complex_count = subunits.groupby([
'dataset', 'dataset_name', 'condition', 'growth_rate_hr',
'complex_annotation', 'complex'
])['n_units'].mean().reset_index()
complex_ribo = complex_count[complex_count.complex_annotation == 'ribosome']
for g, d in complex_ribo.groupby(['dataset', 'dataset_name']):
ax.plot(size.lambda2size(d['growth_rate_hr']),
d['n_units'],
'o',
color=dataset_colors[g[0]],
alpha=0.75,
markeredgecolor='k',
markeredgewidth=0.25,
label=g[1],
ms=4,
zorder=10)
ax.set_xlabel('estimated cell volume [fL]', fontsize=6)
ax.set_ylabel('ribosomes per cell', fontsize=6)
ax.xaxis.set_tick_params(labelsize=5)
ax.yaxis.set_tick_params(labelsize=5)
ax.legend(fontsize=6, loc='upper left')
reported_volume = rates.loc[rates['condition'] ==
c]['volume_fL'].values[0]
gene_dict = {
'gene_name': g,
'b_number': b_number,
'condition': c,
'reported_tot_per_cell': d[f'{c}_tot'].values[0],
'reported_fg_per_cell': d[f'{c}_tot'].values[0] * mw,
'go_terms': go_term,
'cog_class': cog_class,
'cog_category': cog_cat,
'cog_letter': cog_letter,
'growth_rate_hr': growth_rate,
'gene_product': gene_product,
'reported_volume': reported_volume,
'corrected_volume': size.lambda2size(growth_rate)
}
dfs.append(pd.DataFrame(gene_dict, index=[0]))
else:
print(f'Warning!!! {g} not found in the gene list!')
for c in conditions:
growth_rate = rates.loc[rates['condition'] ==
c]['growth_rate_hr'].values[0]
reported_volume = rates.loc[rates['condition'] ==
c]['volume_fL'].values[0]
gene_dict = {
'gene_name': g[0],
'b_number': g[1],
'condition': c,
'reported_tot_per_cell': d[f'{c}_tot'].values[0],
'reported_fg_per_cell': d[f'{c}_fg'].values[0],
label=g[1],
ms=4,
zorder=10)
ax[0].set_xlabel('estimated # ori', fontsize=6)
ax[0].set_ylabel('ribosomes per cell', fontsize=6)
ax[0].xaxis.set_tick_params(labelsize=5)
ax[0].yaxis.set_tick_params(labelsize=5)
ax[0].legend(fontsize=6, loc='upper left')
print(complex_ribo['n_units'].max() / complex_ribo['n_units'].min())
## Plot of ribosome concentration
for g, d in complex_ribo.groupby(['dataset', 'dataset_name']):
ax[1].plot(d['growth_rate_hr'],
d['n_units'] / size.lambda2size(d['growth_rate_hr']),
'o',
color=dataset_colors[g[0]],
alpha=0.75,
markeredgecolor='k',
markeredgewidth=0.25,
label=g[1],
ms=4,
zorder=10)
ax[1].set_xlabel('growth rate [hr$^{-1}$]', fontsize=6)
ax[1].set_ylabel('ribosome concentration [fL$^{-1}$]', fontsize=6)
ax[1].xaxis.set_tick_params(labelsize=5)
ax[1].yaxis.set_tick_params(labelsize=5)
ax[1].set_ylim(0, 40000)
# ax[1].legend(fontsize=6, loc = 'upper left')
